Open crunchyintheory opened 8 years ago
Just to give one more example besides Github:
Any progress on this? Flagging so many safe messages as suspicious will cause people to care less about the warning, undermining its purpose.
I'd like to be able to tell N1 that X email is NOT a problem, and help it learn. How do I do that?
It flags pretty much any email sent from an automated system, maybe implement a rule that checks for if it's a subdomain of the current domain, or has 'reply' in it or something.